Aleksei Zinovev

Hi @kokorins happy to see you again First of all, I need to share with you the original [model](https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov4) and pre and post-processing for it. Looks like you use the...

Great summarization of a few proposals that have been discussed in the past with you. So I suggest arranging a meeting in October with other community members to discuss it....

Yes, a lot of tests are missed for many layers. It will be great to add them to the test suite

Yes, put it into one file

Thanks for the note @therealansh ! Probably the `Average` layer could be implemented with a bug (low probability), or you could find something else (high probability) that influences on the...

Keep in mind the predictSoftly method for the TensorFlow models before refactoring The initial idea of the predictSoftly method is the following: return the vector of probabilities instead of the...

Thanks to the issue and proposed PR for this issue, we need to design in this place a better architecture to add new data sources more easily than now. It...

@AlexandreBrown Unfortunately, it was not included

First of all, thanks for spending time on this benchmark! It gives us food for thinking. In both cases, it doesn't depend on Kotlin or Python. In both cases, it...

Hi @wuhanstudio , could I suggest you check the following scenario, based on deeper VGG network and bigger dataset, cifar10 (it also embedded dataset in [Keras](https://keras.io/api/datasets/cifar10/)) The basic code for...